blueqat/blueqatSDK

What it solves

blueqatSDK provides a comprehensive toolkit for designing, simulating, and optimizing quantum circuits. It bridges the gap between high-level quantum algorithm design and low-level hardware implementation, supporting both standard gate-based quantum computing and specialized exchange-only spin qubit hardware.

How it works

Built on PyTorch, the SDK offers two primary simulation modes: a dense statevector simulator for small circuits and a memory-scalable tensor-network (tensornet) simulator for larger ones. Because it is PyTorch-based, the circuits are differentiable, allowing for gradient-based optimization of circuit parameters. It supports a wide range of features including noise modeling (depolarizing, amplitude damping, and quasi-static noise), error correction experiments, and Clifford operator stabilization.

Who it’s for

Quantum computing researchers, developers, and physicists who need to simulate quantum algorithms, perform Variational Quantum Eigensolver (VQE) or QAOA optimizations, or target specific silicon spin qubit hardware.

Highlights

• Differentiable Simulation: Leverages PyTorch to maintain gradients through circuit execution for parameter optimization.
• Hybrid Simulation Modes: Switch between statevector and tensor-network contraction based on memory requirements.
• Noise & Error Modeling: Includes density-matrix simulation, various noise channels, and tools for testing quantum error correction codes.
• Hardware Specialization: Dedicated support for exchange-only spin qubits, including pulse synthesis and transpilation.
• LLM Integration: Provides an MCP server allowing AI agents (like Claude) to run circuits and analyze quantum results.
• Cloud & Hardware Access: Integration with a cloud API for remote simulation and submission to real QPU hardware.